Balkinization: Why voting structures matter

Why voting structures matter

Sandy Levinson

The New York Times reports that Democrats are "worried" about their 33-year-old candidate for the Senate in Illinois, who won the primary on Tuesday with a ringing 39% of the vote. For all I know he will be a terrific candidate and a wonderful senator (if elected), but what can be said beyond reasonable doubt (as Republicans are apparently already saying) is that 61% of the Democratic electorate preferred someone else. Perhaps Alexei Giannoulias would have prevailed in the runoff, but, obviously we'll never know. But what I think we can know is that runoff systems, though not perfect, are more likely to produce a candidate acceptable to the majority (which some people say is what "democracy" means) than first-past-the-post systems like that in Illinois (or in Texas, which re-elected Gov. Rick Perry in 2006 with a whopping 37% of the statewide vote). If the Democrats manage to lose the seat in November, they might consider blaming an election system that is quite literally designed to stymie "majority rule" save in races where there are only two candidates. (There were three plausible candidates in the Democratic primary, including an attractive "reformer" who received about 33% of the vote, just as there were four such candidates (or at least 3-1/2, depending on what you think of Kinky Friedman's candidacy) in Texas in 2006.

Posted 11:57 AM by Sandy Levinson [link]

Comments:

Sandy:

If the Democrats manage to lose the seat in November, they might consider blaming an election system that is quite literally designed to stymie "majority rule" save in races where there are only two candidates. (There were three plausible candidates in the Democratic primary, including an attractive "reformer" who received about 33% of the vote, just as there were four such candidates (or at least 3-1/2, depending on what you think of Kinky Friedman's candidacy) in Texas in 2006.

I like the idea of runoff general elections so third parties do not act to split majorities ala NY-23. However, I have not seen any substantive evidence that primary runoffs generate more popular candidates.

In any case, if both parties are under the same electoral "handicap," the Dems are at no electoral system disadvantage. Rather, they are already making pre-emptive excuses for losing the Obama seat in what appears to be more of a 1946 than a 1994 level wave election.

When the GOP is actually leading in generic congressional polling of adults done by Dem pollsters and have broken to bigger leads among registered and likely voters, that is a once in a generation tsunami warning. The GOP did not have this kind of lead in 1994. Every single Dem with GOP competition in a district or state with less than 50% Dem registered voters is in serious trouble.

# posted by

Bart DePalma : 6:03 PM

Given Bart's track record from the 2006 and 2008 elections, including the impeccable logic with which he explained the certainty of McCain's pending victory, any Democrat would have to be encouraged by reading his last comment.

# posted by

Steve M : 6:25 PM

Its too late to send the college more money to put on McCain or I would definitely send them another $100.

These exit polls are fantastic!

# posted by Bart DePalma : 6:06 PM

My all-time favorite Baghdad Bart post.

# posted by

Bartbuster : 6:59 PM

Is there some evidence to suggest that a different candidate would have won if the field were narrowed to the top two vote-getters? I am not sure what you mean by blaming the "election system." Is there anything that prohibits the Illinois Democratic Party from having a runoff primary if it wishes? Presumably it would do so if it thought that such a system would often produce a different, and more electable, candidate. On the other hand, one could see that such a system would have substantial costs, not only in terms of money but in terms of increasing divisions in the party. And there is no guarantee that a runoff system would produce the "best" candidate in terms of uniting the party and winning the election. It might just favor the candidate with the most resources or the most committed followers (willing to vote in multiple primaries).

My guess is that the party believes that a runoff system's costs would outweigh the benefits.

# posted by

mls : 10:57 PM

mls, I'm pretty sure Prof. Levinson is referring to instant runoff voting.

# posted by

Mark Field : 10:42 AM

The plurality system is about the worst method there is to elect a candidate. If the field is split by enough candidates one could win with only 1% of the vote, or even just 1 vote.

Instant runoff voting and other runoff methods are a big improvement but still flawed. Many of us who study voting systems consider Condorcet methods the most democratic.

(http://en.wikipedia.org/wiki/Condorcet_method)

Under a Condorcet method voters rank the candidates in order of preference (1,2,3..) and the winner is simply the candidate that a majority prefer to any other candidate.

The problem with runoff voting is often it eliminates this "ideal" candidate.

# posted by

Unknown : 11:32 AM

Yes, we know the problems when some guy from Illinois received 39.8% of the popular vote a while back.

I think IRR should be tried more. But, perhaps germane topic: consider the expanded list of "best picture" nominations. Some which simply are not that good. The winner can have quite a small percentage.

# posted by

Joe : 11:35 AM

Condorcet Method, which needs to get some more press, does sound interesting.

# posted by

Joe : 11:49 AM

Under a Condorcet method voters rank the candidates in order of preference (1,2,3..) and the winner is simply the candidate that a majority prefer to any other candidate.

Ideal? Except for Arrow's Theorem.

# posted by

Mark Field : 12:35 PM

Just a quick comment on Prof. Balkin's post above. I see no reasonable liklihood that the Democrats would change the existing Senate rules. I do believe that if the Republicans were to regain the Senate, and the Democratic minority were to pull the same stunt Shelby has, then the rules would be changed.

# posted by

Mark Field : 3:11 PM

Good news, the Democrats are no longer that worried about their Senate nominee:

"Scott Lee Cohen (Ill.), the Democratic lieutenant governor nominee, is considering dropping his bid after opposition to his candidacy has risen, a Chicago television station reported Saturday.

Cohen's ex-wife has accused him of taking steroids and threatening her and her children. Cohen reportedly attempted to choke his ex-wife Debra."

# posted by

mls : 10:05 AM

Politics in Illinois is a freak show. Just when you think you've seen the worst of it, something comes up that broadens your understanding of the word "corruption" and its many applications.

# posted by

PMS_CC : 10:20 AM

Just a reminder that majority vote requirements have been used to prevent the election of racial minorities. Jesse Jackson was all over this issue some years ago.

Illinois Democrats really should change their system of selecting Lt. Gov and Gov. nominees independently in the primary when they have to run together in the general.

# posted by

John Tanner : 12:11 PM

I don't think Dem's were worried. Alexei would have lost in the runoff anyways...although Rick was re-elected in Texas, he is struggling to get into Washington.

~DREW
what is the bible?

# posted by

KB : 12:44 PM

"Just a reminder that majority vote requirements have been used to prevent the election of racial minorities. Jesse Jackson was all over this issue some years ago."

The problem with the plurality system is that it can accidentally elect a candidate opposed by a majority.

This means that if a majority are racists then a more democratic voting system may inadvertantly lead to less African Americans being elected.

The solution to this is to move to proportional representation, not to keep the terrible plurality system. Occasionally 'splitting' the vote of a racist majority isn't really the best way for minorities to be elected.

# posted by

Unknown : 3:45 PM

# posted by

Unknown : 1:12 PM